With few exceptions, extensive characterization of signal pathway alterations in cancer, and advances in our capacity to identify potent modulators of distinct molecular targets, has not translated into breakthrough drugs. In a recent New York Times article (Nov. 14, 2009), Mervyn Turner, Chief Strategy Officer for Merck was quoted: We invest far too long in bad ideas. It is really important to stop that at an earlier stage in the cycle. This is particularly evident in cancer drug development where over 90% of new anti-cancer drugs fail in the clinic. Our experiences in this area lead us to believe that one reason fo this problem is the lack of an efficient, clinically predictive platform for validating hypotheses nd making program advancement decisions at the earliest stages of the drug development process. Current in vivo models are resource intensive and rate limiting. Furthermore, current in vivo drug efficacy testing requires substantial investment in chemistry to optimize compounds for systemic delivery. Thus the majority of early drug development is heavily dependent on in vitro tumor cell models most of which are grown under conditions, such as in serum, which do not represent the native tumor environment. It is well accepted that these models do not reliably predict clinical efficacy. A second fundamental issue is that almost all tumors acquire resistance to single agent targeted therapies by rewiring signaling networks to reestablish effective oncogenic output. Proactive and rapid identification of alternative druggable pathway nodes would enable a shift in focus from developing drugs that target specific molecules to devising strategies to shut down oncogenic networks. To address these issues, we are developing an end-to-end platform that enables comprehensive functional interrogation of cancer pathways of interest to drug developers, compatible with rapid validation of findings in vivo co-clinical model at the earliest stages of cancer drug development. In Aim 1, we will employ a gene trap vector to generate tumor models that emit light only upon specific inhibition of key oncogenic pathway nodes. In Aim 2, we will use our porous needle array technology to demonstrate capacity to test, compare, and validate multiple agents, in the context of a single tumor grown in vivo, for anti-pathway and anti-tumor efficacy. Drs. Klinghoffer and Olson (Presage Biosciences Inc., and the Fred Hutchinson Cancer Research Center), who developed the porous needle array technology, have joined forces with Dr. Finney (Xactagen LLC), whose optimized gene trap vector system permits capture of pathway modulation via endogenous gene regulation with sensitivity amenable to in vivo analysis. Successful coupling of the porous needle array with high fidelity, high sensitivity gene trap derived oncogene pathway reporters will facilitate a paradigm shift from the current heavy reliance on standard in vitro based models early in the drug discovery process to more predictive and disease relevant in vivo models. Our platform will improve rational advancement of drug development projects, and in doing so, will increase the clinical success rate of new treatments that improve the health of cancer patients.